Public health needs epidemiology heatmaps by ZIP3 while hiding patient ZIP+4 detail from general analysts. What should they configure?

Applying a column mask that conceals ZIP+4 while leaving ZIP3 visible for heatmaps is the correct Unity Catalog control for public health. The mask redacts the finer-grained patient geography for general analysts while preserving coarser ZIP3 needed for epidemiology heatmaps, balancing civic privacy with analytical utility on one governed table.

Enabling transparent data encryption for ZIP+4 while keeping ZIP3 readable for heatmaps fails because it encrypts data at rest but decrypts transparently for queries, so general analysts would still see ZIP+4; it does not govern per-principal column-level visibility at all. Defining a row-level security policy on ZIP+4 while keeping ZIP3 visible for heatmaps fails because it filters which patient rows are returned, not the ZIP+4 values inside those rows, so it cannot conceal finer geography while leaving ZIP3 exposed for heatmaps. Applying a display format to ZIP+4 showing ZIP3 while storing raw ZIP+4 for heatmaps fails because it only changes presentation without restricting access, so general analysts can still query and read the raw ZIP+4 the stem requires hidden.

Exam caveat: Column masks can expose different transformations to different principals—general analysts see ZIP3-only or masked ZIP+4, while authorized epidemiologists might retain finer fields if policy allows. Masks are not row filters; they do not remove patient rows, only reshape column values. ZIP3 heatmaps can still present re-identification risk in sparse cells; pair masking with aggregation standards where needed.

Operational check: Configure the mask so general analysts cannot read ZIP+4 but can use ZIP3, render a sample heatmap on ZIP3, query as a general analyst and confirm ZIP+4 is concealed, and confirm that applying a display format alone does not mask ZIP+4 for general analysts.